Paper on semantics of passwords presented at NDSS, San Diego

On February 25, Rafael Veras presented the results of his Masters research at the Network and Distributed Systems Security Symposium 2014, which took place in San Diego. The 15-page article, co-authored by Julie Thorpe and Christopher Collins is the first to show how semantic patterns extracted from publicly available password leaks can significantly improve the guessing performance of off-line attacks, exposing a security vulnerability. Rafael continues as a member of the vialab, currently in the first year of his PhD studies. He has been studying how data-centric measures of interestingness can be used to suggest relevant views of the data in visual analytics applications. Check out the UOIT news article describing this work.